IBM i Mobile Native Apps Coming On Strong

July 16, 2012 Dan Burger

You know you have business apps and data running on the IBM i operating system that needs to be mobilized. The business case for this is obvious. It solves a common business problem. That is, if you don’t have slick-running business apps for a mobile device, you have a problem. Real time access to information for your mobile workforce is a revenue boosting tool that is too powerful to ignore. But there’s still a decision to be made: Do you want apps running native on the mobile device or do you tweak Web apps for mobile uses?

In the early going, the consensus opinion was to develop Web apps that would render on multiple mobile devices. For those already doing Web development work, this was a natural progression. Developing applications for Apple‘s iOS or Google‘s Android operating systems was not a skill that many IBM midrange shops possessed. And where the skills were present–like a roomful of Java programmers hungry to build Android apps–there was the dominance of the Apple iPad that gives rise to second thoughts. Then again, finding programmers at midsize companies who can pump out Objective C applications for iPads is about as unlikely as finding cows that squirt out chocolate milk.

But now we are seeing software vendors offering frameworks that make it easier for developers to create native applications specifically built so RPG apps can be extended to run natively on the iPad, with other mobile device coverage coming soon. These frameworks are not unique to RPG. They exist for other programming languages as well. But for now, let’s talk like mobile application development in an RPG world.

LANSA, the IBM i application development tool vendor, created a hive of activity around its booth at the COMMON Annual Meeting and Exposition two months ago when it announced the availability of LongRange, an intermediate server and a iPad app that together convert RPG code and DDS displays so applications can run natively on that device. Steve Gapp, LANSA’s CEO, told me last week, a version of LongRange for LANSA’s 4GL customers that want to create native iPad apps has just been released. He also said an RPG version of the product capable of creating native apps on Android is imminent. Farther down the road, he says LANSA will have an app designed for Windows 8 tablet users.

At that point, Gapp says, users can write an application once in RPG and DDS, integrate it with LongRange, and have the potential to deploy the app on the operating systems of the major tablet makers. End users need only to buy the app for the tablet they are using.

The importance of LongRange is that it only requires RPG programmers to know RPG. Skills in other programming languages are unnecessary. It does require the writing of new RPG and DDS code that acts as the user interface layer. Templates are built into LongRange to assist RPG programmers with the creation of such features as drop downs, text boxes, buttons, time widgets, prompters, and many other features that take advantage of the built-in capabilities of an iPad. It’s only by running native on the iPad that technology Apple puts into its market-hogging tablet can be accessed. Web-based applications can forget about that.

Red Oak Software, a terminal emulation vendor in the IBM i and the System z mainframe markets, is also ready to lend a hand to RPG programmers looking for an easier way to get native apps on the iPad. Red Oak’s app, called Legacy Mobile, is designed to display green-screen applications. Coming from years of experience in the emulation business, this seems like a smart way to break into this business. Red Oak CEO George Cummings calls it “a starting point, not a goal.”

Cummings describes the base product as “an intelligent emulator” that passes data back and forth with the capability to execute green-screens as fast as anyone can. For users who are familiar with maneuvering through this territory–like Red Oak customers would be–there are productivity benefits that non-green-screen users would probably never understand. The iPad interface, naturally, includes the keyboard and special function keys that make sense for these users. Red Oaks existing customers have all received a trial version of Legacy Mobile.

Since its founding Red Oak’s development of “programmatic integration” tools–products, that intercept screen definition languages, like the 5250 data stream, and transform them into Windows or Web browser GUIs–in Java. That language would point mobile development in the direction of Android-based devices. But Cummings said the decision regarding its mobile product lines were made a year ago and it was based on business computing being standardized on the iPad.

Expanding on that, Cummings says, “We don’t care what the underlying app is written in,” Cummings says. “As long as the functionality can be accessed with 5250, we can help.”

Leveraging existing skill sets into mobile solutions like LANSA and Red Oak are doing is a great idea, says Bill Gravelle, an independent consultant with experience in mobility, modernization, and integration projects. He prefers the native app engine approach over the Web application strategy, which he describes as focused on “fitting a browser page, made for a 24-inch monitor, into the small glass.”

But he is also wary of some of the native app engine blueprints.

“It’s a step in the right direction for a certain set of business developers with a certain set of mobile solution requirements,” he says. “I just think there are better solutions for broader audiences.”

Productivity is the goal and learning curves are the bumps in the road. In the case of RPG developers, the avoidance of retraining in other languages and technologies keeps the learning curve relatively flat. This is not to say RPG programmers are any different than any other programmers. It’s just to keep the conversation in the RPG world.

“There are many frameworks that promote themselves in a similar manner, but their learning curve takes the traditional business developer into traditional coding. In my mind, that’s not moving the ball far enough down the field,” Gravelle says. “I also think that, for the most part, these frameworks lead to ‘quick and dirty’ solutions that generally fall short in the real-world of user experience.”

Getting a few simple applications up and running on a mobile device is often the early goal of a long-term, multiple phased project. It’s instant gratification, if you want to look at it that way. But it’s also taking a learn-as-you-go, don’t-jump-in-until-you-find-out-how-deep-the-water-is approach.

Gravelle sees it differently. He believes with better tooling and better skills, it’s possible to deliver apps that make a real difference to mobile knowledge workers in a much shorter time frame.

The issue goes beyond the initial decision of whether to choose a native app or browser solution, even though Gravelle leans toward native apps because of their better user experience.

“The difference-maker in our competitive world is the choice of tooling/platforms. It’s my belief that continuing traditional coding approaches is counter-productive for enterprise mobile solutions–the issue of Java versus Objective-C versus HTML5/CSS3 is really not the major one,” he says. “I’m probably one of few out there these days who believes that the less code I have to personally develop to quickly deliver an effective solution actually makes me a more valuable developer. As we in the IBM i world have known for years, building business solutions on top of comprehensive, productive, integrated, and innovative platforms actually delivers a tremendous competitive advantage for the enterprise.”

“Less code doesn’t mean no logic, no process, no workflow–it just means that no traditional code is necessary (no RPG, no Java, no Objective-C, etc). It means that the existing developers do tend to elevate into architects or conductors who master the orchestration of these pre-built components or templates with configuration options.”